1. DCSP-16 Jianfeng Feng Department of Computer Science Warwick Univ., UK Jianfeng.feng@warwick.ac.uk http://www.dcs.warwick.ac.uk/~feng/dsp.html

2. Applications of filter

3. Example Right hemisphere Left hemisphere

4. HumanEar.jpg Frequency band [f1 f2] Frequency band [f3 f4]

5. simple filter design In a number of cases, we can design a linear filter almost by inspection, by moving poles and zeros like pieces on a chessboard. This is not just a simple exercise designed for an introductory course, for in many cases the use of more sophisticated techniques might not yield a significant improvement in performance.

6. In this example consider the audio signal s(n), digitized with a sampling frequency Fs=12kHz. The signal is affected by a narrowband (i.e., very close to sinusoidal) disturbance w(n). Fig. show the frequency spectrum of the overall signal plus noise, x(n)=s(n)+w(n), which can be easily determined.

8. Notice two facts: The signal has a frequency spectrum concentrated within the interval 0 to 6 kHz. The disturbance is at frequency F 0 =1.5 kHz. Now the goal is to design and implement a simple filter that rejects the disturbance without affecting the signal excessively. We will follow three steps:

9. Step 1: Frequency domain specifications We need to reject the signal at the frequency of the disturbance. Ideally, we would like to have the following frequency response: where 0 =2 (F 0 /F s )= /4 radians, the digital frequency of the disturbance.

10. Step 2: Determine poles and zeros

11. /4

16. Step 3: Determine the difference equation in the time domain From the transfer function, the difference equation is determined by inspection: y(n)=1.343 y(n-1)-0.9025 y(n-2)+0.954 x(n) - 1.3495x(n-1)+0.9543 x(n-2)

17. The difference equation can be easily implemented as a recursion in a high-level language. The final result is the signal y(n) with the frequency spectrum shown in Fig., where we notice the absence of the disturbance.

19. demo C:\Program Files\MATLAB71\work simple_filter_design soundsc(s,Fs) soundsc(x,Fs) soundsc(y,Fs) soundsc(yrc,Fs)